Fake News Detection System, based on CBOW and BERT

Abstract

Fake news is becoming a major challenge that greatly affects the public’s trust in the media. In this paper, we propose a new solution, combining word embedding based on CBOW (Continuous Bag Of Words) and the BERT (Bidirectional Encoder Representations from Transformers) models to support fake news detection. This paper focuses on presenting the proposed model and processing steps through the FND4Vn system, with a data set of Vietnamese news. Experimental results show that this solution achieves accuracy as high as 0.96 in recall and has many advantages compared to existing methods.